oracle 🧿 — Whispering your tokens to the silicon sage

Oracle bundles your prompt and files so another AI can answer with real context. It speaks GPT-5.5 Pro (default), GPT-5.5, GPT-5.4 Pro, GPT-5.4, GPT-5.1 Pro, GPT-5.1 Codex (API-only), GPT-5.1, GPT-5.2, Gemini 3.1 Pro (API-only), Gemini 3 Pro, Claude Sonnet 4.6, Claude Opus 4.1, and more—and it can ask one or multiple models in a single run. Browser automation is available; use --browser-model-strategy current to keep the active ChatGPT model (or ignore to skip the picker). API remains the most reliable path, and --copy is an easy manual fallback.

Setting up (macOS Browser Mode)

Browser mode lets you use GPT-5.5 Pro without any API keys — it automates your Chrome browser directly.

First-time login

Run this once to create an automation profile and log into ChatGPT. The browser will stay open so you can complete the login:

oracle --engine browser --browser-manual-login \
  --browser-keep-browser --browser-input-timeout 120000 \
  -p "HI"

Subsequent runs

Once logged in, the automation profile is saved. Use this for all future runs:

oracle --engine browser --browser-manual-login \
  --browser-auto-reattach-delay 5s \
  --browser-auto-reattach-interval 3s \
  --browser-auto-reattach-timeout 60s \
  -p "your prompt"

Why these flags?

• --browser-manual-login — Skips macOS Keychain cookie access (avoids repeated permission popups)
• --browser-auto-reattach-* — Reconnects when ChatGPT redirects mid-page-load (fixes "Inspected target navigated or closed" error)
• --browser-keep-browser — Keeps browser open for first-time login (not needed after)
• --browser-input-timeout 120000 — Gives you 2 minutes to log in on first run

Quick start

Install globally: npm install -g @steipete/oracle Homebrew: brew install steipete/tap/oracle

Requires Node 22+. Or use npx -y @steipete/oracle … (or pnpx).

# Copy the bundle and paste into ChatGPT
npx -y @steipete/oracle --render --copy -p "Review the TS data layer for schema drift" --file "src/**/*.ts,*/*.test.ts"

# Minimal API run (expects OPENAI_API_KEY in your env)
npx -y @steipete/oracle -p "Write a concise architecture note for the storage adapters" --file src/storage/README.md

# Multi-model API run
npx -y @steipete/oracle -p "Cross-check the data layer assumptions" --models gpt-5.1-pro,gemini-3-pro --file "src/**/*.ts"

# Follow up from an existing OpenAI/Azure session id
npx -y @steipete/oracle --engine api --model gpt-5.2-pro --followup release-readiness-audit --followup-model gpt-5.2-pro -p "Re-evaluate with this new context" --file "src/**/*.ts"

# Follow up directly from an OpenAI Responses API id
npx -y @steipete/oracle --engine api --model gpt-5.2-pro --followup resp_abc1234567890 -p "Continue from this response" --file docs/notes.md

# Preview without spending tokens
npx -y @steipete/oracle --dry-run summary -p "Check release notes" --file docs/release-notes.md

# Browser run (no API key, will open ChatGPT)
npx -y @steipete/oracle --engine browser -p "Walk through the UI smoke test" --file "src/**/*.ts"

# Browser multi-turn consult in one ChatGPT conversation
npx -y @steipete/oracle --engine browser --model gpt-5.5-pro \
  -p "Review this migration plan" --file docs/migration.md \
  --browser-follow-up "Challenge your previous recommendation" \
  --browser-follow-up "Give the final decision"

# Gemini browser mode (no API key; uses Chrome cookies from gemini.google.com)
npx -y @steipete/oracle --engine browser --model gemini-3-pro --prompt "a cute robot holding a banana" --generate-image out.jpg --aspect 1:1

# Sessions (list and replay)
npx -y @steipete/oracle status --hours 72
npx -y @steipete/oracle session <id> --render
npx -y @steipete/oracle restart <id>

# TUI (interactive, only for humans)
npx -y @steipete/oracle tui

Engine auto-picks API when OPENAI_API_KEY is set, otherwise browser; browser is stable on macOS and works on Linux and Windows. On Linux pass --browser-chrome-path/--browser-cookie-path if detection fails; on Windows prefer --browser-manual-login or inline cookies if decryption is blocked.

Integration

CLI

• API mode expects API keys in your environment: OPENAI_API_KEY (GPT-5.x), GEMINI_API_KEY (Gemini 3.1 Pro / Gemini 3 Pro), ANTHROPIC_API_KEY (Claude Sonnet 4.6 / Opus 4.1).
• Gemini browser mode uses Chrome cookies instead of an API key—just be logged into gemini.google.com in Chrome (no Python/venv required).
• If your Gemini account can’t access “Pro”, Oracle auto-falls back to a supported model for web runs (and logs the fallback in verbose mode).
• Prefer API mode or --copy + manual paste; browser automation is experimental.
• Browser support: stable on macOS; works on Linux (add --browser-chrome-path/--browser-cookie-path when needed) and Windows (manual-login or inline cookies recommended when app-bound cookies block decryption).
• Remote browser service: oracle serve on a signed-in host; clients use --remote-host/--remote-token.
• Browser artifacts: browser sessions save transcript.md and generated artifacts under ~/.oracle/sessions/<id>/artifacts/. Deep Research saves deep-research-report.md when the report surface is captured; ChatGPT-generated images are downloaded with the active browser cookies when image URLs are present.
• AGENTS.md/CLAUDE.md:

  - Oracle bundles a prompt plus the right files so another AI (GPT 5 Pro + more) can answer. Use when stuck/bugs/reviewing.
  - Run `npx -y @steipete/oracle --help` once per session before first use.
  

• Tip: set browser.chatgptUrl in config (or --chatgpt-url) to a dedicated ChatGPT project folder so browser runs don’t clutter your main history.

Codex skill

• Copy the bundled skill from this repo to your Codex skills folder:
  • mkdir -p ~/.codex/skills
  • cp -R skills/oracle ~/.codex/skills/oracle
• Then reference it in your AGENTS.md/CLAUDE.md so Codex loads it.

MCP

• Run the stdio server via oracle-mcp.
• Configure clients via steipete/mcporter or .mcp.json; see docs/mcp.md for connection examples.
• Claude Code on the same Mac as a signed-in ChatGPT browser can generate a local config directly:

oracle bridge claude-config --local-browser > .mcp.json

• In MCP consult, use preset: "chatgpt-pro-heavy" for ChatGPT browser mode with gpt-5.5-pro and heavy thinking. Add dryRun: true to inspect the resolved run without creating a session or touching Chrome.

npx -y @steipete/oracle oracle-mcp

• Cursor setup (MCP): drop a .cursor/mcp.json like below, then pick “oracle” in Cursor’s MCP sources. See https://cursor.com/docs/context/mcp for UI steps.

{
  "oracle": {
    "command": "oracle-mcp",
    "args": []
  }
}

Highlights

• Bundle once, reuse anywhere (API or experimental browser).
• Multi-model API runs with aggregated cost/usage, including OpenRouter IDs alongside first-party models.
• Claude Code / MCP browser consults can use the chatgpt-pro-heavy preset for a compact ChatGPT Pro second-opinion workflow.
• Render/copy bundles for manual paste into ChatGPT when automation is blocked.
• GPT‑5 Pro API runs detach by default; reattach via oracle session <id> / oracle status or block with --wait.
• OpenAI/Azure follow-up API runs can continue from --followup <sessionId|responseId>; for multi-model parents, add --followup-model <model>.
• Azure endpoints supported via --azure-endpoint/--azure-deployment/--azure-api-version or AZURE_OPENAI_* envs.
• File safety: globs/excludes, size guards, --files-report.
• Sessions you can replay (oracle status, oracle session <id> --render).
• Session logs and bundles live in ~/.oracle/sessions (override with ORACLE_HOME_DIR).

Follow-up and lineage

Use --followup to continue an existing OpenAI/Azure Responses API run with additional context/files:

oracle \
  --engine api \
  --model gpt-5.2-pro \
  --followup <existing-session-id-or-resp_id> \
  --followup-model gpt-5.2-pro \
  --slug "my-followup-run" \
  --wait \
  -p "Follow-up: re-evaluate the previous recommendation with the attached files." \
  --file "server/src/strategy/plan.ts" \
  --file "server/src/strategy/executor.ts"

When the parent session used --models, --followup-model picks which model's response id to chain from. Custom --base-url providers plus Gemini/Claude API runs are excluded here because they do not preserve previous_response_id in Oracle.

oracle status shows parent/child lineage in tree form:

Recent Sessions
Status    Model         Mode    Timestamp           Chars    Cost  Slug
completed gpt-5.2-pro   api     03/01/2026 09:00 AM  1800  $2.110  architecture-review-parent
completed gpt-5.2-pro   api     03/01/2026 09:14 AM  2200  $2.980  ├─ architecture-review-followup
running   gpt-5.2-pro   api     03/01/2026 09:22 AM  1400       -  │  └─ architecture-review-implementation-pass
pending   gpt-5.2-pro   api     03/01/2026 09:25 AM   900       -  └─ architecture-review-risk-check

Browser auto-reattach (long Pro runs)

When browser runs time out (common with long GPT‑5.x Pro responses), Oracle can keep polling the existing ChatGPT tab and capture the final answer without manual oracle session <id> commands.

Enable auto-reattach by setting a non-zero interval:

• --browser-auto-reattach-delay — wait before the first retry (e.g. 30s)
• --browser-auto-reattach-interval — how often to retry (e.g. 2m)
• --browser-auto-reattach-timeout — per-attempt budget (default 2m)

oracle --engine browser \
  --browser-timeout 6m \
  --browser-auto-reattach-delay 30s \
  --browser-auto-reattach-interval 2m \
  --browser-auto-reattach-timeout 2m \
  -p "Run the long UI audit" --file "src/**/*.ts"

Flags you’ll actually use

Flag	Purpose
-p, --prompt <text>	Required prompt.
-f, --file <paths...>	Attach files/dirs (globs + ! excludes).
-e, --engine <api|browser>	Choose API or browser (browser is experimental).
-m, --model <name>	Built-ins (gpt-5.5-pro default, gpt-5.5, gpt-5.4-pro, gpt-5.4, gpt-5.1-pro, gpt-5-pro, gpt-5.1, gpt-5.1-codex, gpt-5.2, gpt-5.2-instant, gpt-5.2-pro, gemini-3.1-pro API-only, gemini-3-pro, claude-4.6-sonnet, claude-4.1-opus) plus any OpenRouter id (e.g., minimax/minimax-m2, openai/gpt-4o-mini).
--models <list>	Comma-separated API models (mix built-ins and OpenRouter ids) for multi-model runs.
--followup <sessionId|responseId>	Continue an OpenAI/Azure Responses API run from a stored oracle session or resp_... response id.
--followup-model <model>	For multi-model OpenAI/Azure parent sessions, choose which model response to continue from.
--base-url <url>	Point API runs at LiteLLM/Azure/OpenRouter/etc.
--chatgpt-url <url>	Target a ChatGPT workspace/folder (browser).
--browser-model-strategy <select|current|ignore>	Control ChatGPT model selection in browser mode (current keeps the active model; ignore skips the picker).
--browser-manual-login	Skip cookie copy; reuse a persistent automation profile and wait for manual ChatGPT login.
--browser-attach-running	Reuse your current local browser session through local DevToolsActivePort discovery; Oracle opens a dedicated tab instead of launching Chrome (defaults to 127.0.0.1:9222, or combine with --remote-chrome <host:port> to hint a different local endpoint).
--browser-thinking-time <light|standard|extended|heavy>	Set ChatGPT thinking-time intensity (browser; Thinking/Pro models only).
--browser-research deep	Activate ChatGPT Deep Research for broad web research and cited reports (browser only).
--browser-follow-up <prompt>	Browser-only multi-turn consult: submit an additional prompt in the same ChatGPT conversation after the initial answer. Repeat for challenge/revision/final-decision passes. Not supported with Deep Research mode.
--browser-port <port>	Pin the Chrome DevTools port (WSL/Windows firewall helper).
--browser-inline-cookies[(-file)] <payload | path>	Supply cookies without Chrome/Keychain (browser).
--browser-timeout, --browser-input-timeout	Control overall/browser input timeouts (supports h/m/s/ms).
--browser-recheck-delay, --browser-recheck-timeout	Delayed recheck for long Pro runs: wait then retry capture after timeout (supports h/m/s/ms).
--heartbeat <seconds>	Emit API and browser progress heartbeats. Browser mode reports ChatGPT Thinking/Reasoning sidecar liveness metadata when available, without logging reasoning text.
--browser-reuse-wait	Wait for a shared Chrome profile before launching (parallel browser runs).
--browser-profile-lock-timeout	Wait for the shared manual-login profile lock before sending (serializes parallel runs).
--browser-max-concurrent-tabs	Soft limit for simultaneous ChatGPT tabs sharing one manual-login profile (default 3).
--render, --copy	Print and/or copy the assembled markdown bundle.
--wait	Block for background API runs (e.g., GPT‑5.1 Pro) instead of detaching.
--timeout <seconds|auto>	Overall API deadline (auto = 60m for pro, 120s otherwise).
--background, --no-background	Force Responses API background mode (create + retrieve) for API runs.
--http-timeout <ms|s|m|h>	HTTP client timeout (default 20m).
--zombie-timeout <ms|s|m|h>	Override stale-session cutoff used by oracle status.
--zombie-last-activity	Use last log activity to detect stale sessions.
--write-output <path>	Save only the final answer (multi-model adds .<model>). Browser sessions also save transcripts and generated artifacts under ~/.oracle/sessions/<id>/artifacts/.
--files-report	Print per-file token usage.
--dry-run [summary|json|full]	Preview without sending.
--remote-host, --remote-token	Use a remote oracle serve host (browser).
--remote-chrome <host:port>	Attach to an existing remote Chrome session (browser), or when combined with --browser-attach-running use this host:port as the local attach hint.
--youtube <url>	YouTube video URL to analyze (Gemini browser mode).
--generate-image <file>	Generate image and save to file (Gemini browser mode; ChatGPT browser mode saves downloadable image artifacts when present). Extra ChatGPT images save as numbered siblings.
--edit-image <file>	Edit existing image with --output (Gemini browser mode). For ChatGPT browser mode, attach source images with --file and use --generate-image for the output path.
--azure-endpoint, --azure-deployment, --azure-api-version	Target Azure OpenAI endpoints (picks Azure client automatically).

Configuration

Put defaults in ~/.oracle/config.json (JSON5). Example:

{
  model: "gpt-5.5-pro",
  engine: "api",
  filesReport: true,
  browser: {
    chatgptUrl: "https://chatgpt.com/g/g-p-691edc9fec088191b553a35093da1ea8-oracle/project",
  },
}

Use browser.chatgptUrl (or the legacy alias browser.url) to target a specific ChatGPT workspace/folder for browser automation. See docs/configuration.md for precedence and full schema.

When several agents share one manual-login ChatGPT profile, Oracle coordinates browser tab slots through that profile. Extra runs wait and log that they are waiting for a ChatGPT browser slot instead of crashing because another Codex/Claude/CLI run is already using the browser. For the most reliable shared-agent setup, keep one signed-in Chrome open with remote debugging and point callers at it with --remote-chrome <host:port>; direct manual-login launches are guarded so parallel callers reuse the first reachable Chrome instead of racing separate launches on the same profile.

Advanced flags

Area	Flags
Browser	--browser-manual-login, --browser-attach-running, --browser-thinking-time, --browser-research, --browser-follow-up, --browser-timeout, --browser-input-timeout, --browser-recheck-delay, --browser-recheck-timeout, --browser-reuse-wait, --browser-profile-lock-timeout, --browser-max-concurrent-tabs, --browser-auto-reattach-delay, --browser-auto-reattach-interval, --browser-auto-reattach-timeout, --browser-cookie-wait, --browser-inline-cookies[(-file)], --browser-attachments, --browser-inline-files, --browser-bundle-files, --browser-keep-browser, --browser-headless, --browser-hide-window, --browser-no-cookie-sync, --browser-allow-cookie-errors, --browser-chrome-path, --browser-cookie-path, --chatgpt-url
Run control	--background, --no-background, --http-timeout, --zombie-timeout, --zombie-last-activity
Azure/OpenAI	--azure-endpoint, --azure-deployment, --azure-api-version, --base-url

Remote browser example

# Host (signed-in Chrome): launch serve
oracle serve --host 0.0.0.0:9473 --token secret123

# Client: target that host
oracle --engine browser --remote-host 192.168.1.10:9473 --remote-token secret123 -p "Run the UI smoke" --file "src/**/*.ts"

# If cookies can’t sync, pass them inline (JSON/base64)
oracle --engine browser --browser-inline-cookies-file ~/.oracle/cookies.json -p "Run the UI smoke" --file "src/**/*.ts"

Session management

# Prune stored sessions (default path ~/.oracle/sessions; override ORACLE_HOME_DIR)
oracle status --clear --hours 168

More docs

• Bridge (Windows host → Linux client): docs/bridge.md
• Browser mode & forks: docs/browser-mode.md (includes oracle serve remote service), docs/chromium-forks.md, docs/linux.md
• MCP: docs/mcp.md
• OpenAI/Azure/OpenRouter endpoints: docs/openai-endpoints.md, docs/openrouter.md
• Manual smokes: docs/manual-tests.md
• Testing: docs/testing.md

If you’re looking for an even more powerful context-management tool, check out https://repoprompt.com
Name inspired by: https://ampcode.com/news/oracle

More free stuff from steipete

• ✂️ Trimmy — “Paste once, run once.” Flatten multi-line shell snippets so they paste and run.
• 🟦🟩 CodexBar — Keep Codex token windows visible in your macOS menu bar.
• 🧳 MCPorter — TypeScript toolkit + CLI for Model Context Protocol servers.